A method for preparing data for phenotyping wheat seedlings of different varieties using the example of variety "Novosibirskaya 41"

Abstract

The paper discusses the preparation of experimental data used to measure the biopotentials of Novosibirskaya 41 wheat seedlings exposed to elevated and lowered temperatures, in order to conduct phenotyping of these plants using cluster analysis. It is noted that such a preparation is necessary for long-term experimental studies that take several calendar days (up to 10 or more), during which metabolic changes in seedling samples occur, affecting their biopotential values. The paper is based on experimental data obtained in 2020 and 2022 and their regression analysis, as reported in [13]. The results of changes in seedling biopotentials depending on their age are briefly described, and an algorithm for calculating corrective biopotential values for each magnification level of the objects is provided. Statistical regressions of changes in biopotential values depending on the need to preserve seedlings of these wheat varieties were obtained. This allowed the development of an algorithm for correcting the initial average biopotentials for these conditions without preliminary regression analysis of the data. Two data sets were generated for assessing the phenotype of the objects: the original data set, obtained through primary processing of changes in these seedling biopotentials under exposure to elevated and lowered temperatures, and the corrected data set, in the partial parameter (smax.c.) of the maximum filtered centered value (cf) of the wheat seedling biopotentials under these conditions. Plant phenotyping was performed based on the data sets using the original Eclaster program, which implements this methodical spectral clustering from the sklearn.cluster library in the Python programming environment. The clustering results presented in the form of a scatterplot demonstrate improved cluster separation for the corrected data.

Plant phenotyping relevance

小麦幼苗のバイオポテンシャルを用いた表現型評価のため、データ補正アルゴリズムとクラスタリング解析プログラムを開発・適用しており、表現型取得・抽出手法が研究の中心である。

Code and data availability

公開本文の所在を確認できませんでした。非公開または購読が必要な可能性があります。

No evidence-backed public reproduction asset is currently recorded.